Mac OS X: Why does it sometimes refuse to shut down?
Posted by Pierre Igot in: MacintoshFebruary 1st, 2006 • 4:01 pm
This is a problem that, as far as I can tell, is as old as Mac OS X. At least I remember experiencing it occasionally with previous versions of Mac OS X and, since I’ve just experienced it again today, it’s obviously not gone.
In a nutshell, here’s what happens. For some reason, I need to restart your computer. It might be that you’ve just installed a software update that requires it (as I did today) or for some other reason. So you click on the “Restart” button or select the “” command in the Apple menu, and Mac OS X starts quitting all your applications…
Everything goes fine, Mac OS X quits the Finder itself, and the menu bar at the top disappears, with only your desktop background picture or pattern showing. And then the screen turns blue and you get a little “wheel” animation in the middle, indicating the last stages before the screen goes black and the machine actually restarts.
But then Mac OS X gets stuck. Instead of fading to black and actually restarting, Mac OS X stays at the blue screen stage, and the wheel continues to spin endlessly. It’s just happened to me. I waited for a good five minutes, and nothing seemed to be happening. There was no significant hard disk activity that I could tell. Mac OS X just appeared to be stuck.
It might very well be that if I had been patient enough and waited longer, Mac OS X would have finally restarted by itself. But I simply couldn’t wait any longer, so I did a hard reset. Mac OS X booted just fine, and things were normal as far as I could tell.
So what happens exactly when Mac OS X does this? It’s obviously pretty hard to troubleshoot, because it occurs at a stage where the user cannot do anything locally with the machine. It might be possible to use remote login or some other procedure from another machine to see why the system is stuck. I don’t know. I am not enough of an expert in those matters.
And it’s not like it happens all the time. It just happens sometimes. I suppose it’s not a big problem. But it’s annoying just the same, because you don’t know what’s going on and you worry that, if you do a hard reset just after a system update, it might have unfortunate consequences. As far as I can tell, it’s never happened, but you never know. I really don’t like having to do a hard reset unless I absolutely have to. But I cannot afford to wait for 10 or 15 minutes for my machine to restart either.
So I guess it’s one of these problems that is going to persist and continue to occur from time to time for no apparent reason, and I’ll just have to live with it.
February 1st, 2006 at Feb 01, 06 | 4:17 pm
Coincidentallly, this *just* happened to me too (had to turn of machine to clean screen). It’s good to know that it’s still an issue with somebody other than me.
February 1st, 2006 at Feb 01, 06 | 6:05 pm
i’ve had some similarly odd expiences with os x getting to the blue screen and spinner stage after selecting restart or shutdown and then, without restarting, it just reloads the finder/desktop, etc as it it had already restarted.
February 1st, 2006 at Feb 01, 06 | 10:24 pm
I’ve seen it too, almost exactly as described by Pierre, although I don’t know that I’ve ever seen it get stuck and stay stuck… I recall it finally shutting down after a conspicuous delay of perhaps two minutes.
February 3rd, 2006 at Feb 03, 06 | 3:11 am
Hello Pierre, i like your writing style, they really are elaborate and do convey very good what your question is but sometimes a debugging-route is obvious. Have you ever thought that after such things happen and you restart the computer that you might take a look at the console, the system log and what-have-you-not-logs on a unix-like system? Also when a program behaves badly, just open up the console and ‘many’ times it will give hints and notices as what is happening. Success.
February 3rd, 2006 at Feb 03, 06 | 9:29 am
Tr909: The thing to understand is that I am an old-school Mac guy. All the Unix stuff is still rather foreign to me. It’s not “obvious” to me that the solution to my problems might be in some log file with tons of obscure text messages that I have to scroll through manually until I might find a clue regarding why my system behaved the way it did. I am sure you realize that this is not particularly intuitive, and, as such, does not come to mind naturally to a guy like me.
The very fact that you wrote “what-have-you-not-logs” yourself in your comment should give you a clue as to why I didn’t refer to them in my post. There are all kinds of logs in all kinds of locations, and I have no idea which one might be relevant.
And then these log files contain all kinds of information that is obviously totally irrelevant, and I don’t really know what to look for.
That said, I had a look at the system log for the time this particular problem happened, and all I can see is a bunch of lines like this one:
G5 kernel[0]: HFS: /rsrc paths are deprecated (..namedfork/rsrc)Does that tell you anything? It certainly doesn’t tell me anything.
February 4th, 2006 at Feb 04, 06 | 12:10 pm
i am an old skool mac user as wel (i must admit that i started with dos and then windows 3 and came to the mac late 1998). However if you look at the system.log (or crashreporter.log) there are probably a few lines for the very last seconds that might give clues. After that you can just try-out google :-)
Google “G5 kernel[0]: HFS: /rsrc paths are deprecated”
Now, do you use a backup-tool? and does that tool use rsync? (many links google comes up with are RsyncX, CarbonCopyCloner or SilverX related).
February 4th, 2006 at Feb 04, 06 | 12:24 pm
Maybe can you give some more lines and include the time info as well.
A shutdown/reboot command you issue should show up in the log, also when you reset the computer after the following hang should turn up in the log when the machine is booting again after that reset. An entry like ‘starting up system’ or ‘booting started’ or so (i am just as old mac-skool as you, so guessing).
I found that it (unix-like system and logs) is a tremendous help over OS 9.
(hope i can help)
February 5th, 2006 at Feb 05, 06 | 4:04 pm
Tr909: The repeated HFS etc. lines are indeed because of RsyncX. And they actually occurred much later in the day (at the time of my automated backups), so they really don’t have anything to do with the problem at hand.
Here’s an excerpt of my log around the time of the incident with the never-ending shutting down process:
Feb 1 15:19:18 G5 mdimportserver[8952]: *** Failed to decode plain data, treating as binary
Feb 1 15:20:05 G5 crashdump[9000]: crashdump crashed
Feb 1 15:20:05 G5 crashdump[9000]: crash report written to: /Library/Logs/CrashReporter/crashdump.crash.log
Feb 1 15:21:51 G5 mdimportserver[9080]: Missing multipart boundary parameter; using "----==_NextPart_000_05EB_01C61FF7.3B9AC620 "
Feb 1 15:21:51 G5 mdimportserver[9080]: *** Failed to decode quoted-printable data, treating as binary
Feb 1 15:21:51 G5 mdimportserver[9080]: Missing multipart boundary parameter; using "----==_NextPart_000_05EB_01C61FF7.3B9AC620 "
Feb 1 15:23:38 G5 loginwindow[128]: sendQuitEventToApp (LAServer): AESendMessage returned error -609
Feb 1 15:23:38 G5 loginwindow[128]: sendQuitEventToApp (A Better Finder Hot Key): AESendMessage returned error -609
Feb 1 15:23:52 G5 shutdown: reboot by igot:
Feb 1 15:23:53 G5 kernel[0]: jnl: close: flushing the buffer cache (start 0x669c00 end 0x66de00)
Feb 1 15:23:53 G5 SystemStarter[9117]: authentication service (9127) did not complete successfully
Feb 1 15:23:55 G5 SystemStarter[9117]: The following StartupItems failed to properly start:
Feb 1 15:23:55 G5 SystemStarter[9117]: /System/Library/StartupItems/AuthServer
Feb 1 15:23:55 G5 SystemStarter[9117]: - execution of Startup script failed
Feb 1 15:30:01 localhost kernel[0]: standard timeslicing quantum is 10000 us
Feb 1 15:30:01 localhost kernel[0]: vm_page_bootstrap: 1138779 free pages
Feb 1 15:30:00 localhost mDNSResponder-107.4 (Nov 15 2005 21: 34:38)[96]: starting
Feb 1 15:30:01 localhost kernel[0]: mig_table_max_displ = 70
Feb 1 15:30:01 localhost kernel[0]: Copyright (c) 1982, 1986, 1989, 1991, 1993
Feb 1 15:30:01 localhost kernel[0]: The Regents of the University of California. All rights reserved.
Feb 1 15:30:01 localhost kernel[0]: using 8192 buffer headers and 4096 cluster IO buffer headers
Feb 1 15:30:01 localhost kernel[0]: DART enabled
Feb 1 15:30:01 localhost kernel[0]: AppleKauaiATA shasta-ata features enabled
Feb 1 15:30:01 localhost kernel[0]: FireWire (OHCI) Apple ID 52 built-in now active, GUID 001124ff fed6c498; max speed s800.
Feb 1 15:30:00 localhost lookupd[100]: lookupd (version 369.2) starting - Wed Feb 1 15:30:00 2006
Feb 1 15:30:01 localhost kernel[0]: Extension "com.apple.driver.initioFWBridge" has no kernel dependency.
Feb 1 15:30:01 localhost kernel[0]: Extension "com.apple.iokit.IOFireWireSerialBusProtocolTransport" has no kernel dependency.
Feb 1 15:30:01 localhost kernel[0]: Extension "com.apple.driver.IOFireWireSerialBusProtocolSansPhysicalUnit" has no kernel dependency.
Looks like the problem occurred at 15:23 and might have to do with crashdump crashing (!) or the “authentication service,” whatever these things are.